The mechanical stiffness of individual human cells can be a key parameter that reveals dysfunction of the cell. For example, malaria-infected red blood cells are known to be stiffer than uninfected cells and invasive cancer cells can be several times more deformable than healthy phenotypes. However, for biophysical properties to be more useful in biomedical and diagnostic settings, we will require methods for continuous biomechanical fractionation in high throughput, akin to size exclusion chromatography. Although separation by such parameters as size and density are commonly employed, few methods are available for high throughput separation by stiffness and no method for sorting by viscoelasticity. Moreover, because of the potential overlap of biophysical signatures of different cell types, to achieve purity in the separation may require the ability to fractionate cells such that subpopulations can be collected for which the biophysical values do not overlap. Towards these ends, we have created a microfluidic sorting technology that utilizes a combination of hydrodynamic and compressive forces to sort individual cells by biophysical properties. The objective of this research proposal is to create a high-throughput cell fractionation method based on the microfluidic technology that is sensitive to stiffness and viscosity, two orthogonal biophysical phenotypes of cells. The technology consists of a microchannel with periodical, diagonal constrictions that deform cells as they flow to modify their trajectory in a proportion to cell stiffness and viscosity. For example, cells that are stiffer are translated towards the upper part of the channel and cells that are softer migrate towards the bottom part of the channel such that outlets can continuously collect the sorted cells. By engineering channel geometry such as inter-ridge spacing, viscoelastic relaxation of cells can be emphasized, constituting a completely new sorting mechanism not previously utilized. Through computational understanding of channel hydrodynamics and stiffness-dependent trajectories of cells, outlets can be designed to specifically collect the sorted cells and thereby fractionate cells by stiffness. Also, by designin channels with different inter-ridge spacing, differences in cell relaxation rates can be exploited. In preliminary data, we show over 45-fold enrichment of cell types is possible in a label-free manner.